Topology aware functional similarity of protein interaction networks based on gene ontology

Functional comparison and alignment of Protein Interaction Networks (PINs) will enable a better understanding of cellular organization and processes. Gene Ontology (GO) provides a structured standard vocabulary of functional terms of gene products, and has been used to measure the functional similarity between proteins. In this study, we propose an algorithm to measure the functional similarity between PINs based on GO. The algorithm simultaneously takes the PIN's network topology and semantic similarity of the component proteins into account. We employ the algorithm to measure the similarity between pathways present in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database and cluster the pathways according to similarity. The results show great consistency with the function of these pathways.